Snappy : Multilook - GrSquarePixel & AzLooks


I’m new in Snappy (and also in SAR techniques) and I have read this forum to understand a bit how it works and manage to do what I wanted. I’m using Snappy to process Sentinel-1 data, at one point I need to make a Multilooking but I currently have a problem with the number of Azimuth Looks.

When using SNAP desktop to perform Multilook on Sentinel-1 data if the ‘GR Square Pixel’ option is checked, you only have to give the number of Range Looks to have almost square pixels on the resulting image. The software automatically deduced the good number of Azimuth Looks needed to that.

So my question is how can I performed that with Snappy ?

I have tried to put the ‘grSquarePixel’ parameter as ‘True’ and then just give the number of range looks such as :

parameters = snappy.HashMap()
parameters.put(‘grSquarePixel’, True)
parameters.put(‘nRgLooks’, rgLooks)

Where rgLooks is an Integer. But it doesn’t work, I haven’t any error message but the final product is not correct when I make a grayscale jpeg I obtain a fully white image.

To make it works properly I need to add the ‘nAzLooks’ parameter but I cannot let it blank and must set it manually such as :

def multiLooking(product, rgLooks, azLooks, source):

parameters = snappy.HashMap()
parameters.put('grSquarePixel', True)
parameters.put('nRgLooks', rgLooks)
parameters.put('nAzLooks', azLooks)
parameters.put('outputIntensity', False)
parameters.put('sourceBands', source)
multilook_data = snappy.GPF.createProduct('Multilook', parameters, product)
return multilook_data

So how can I obtain an automated calculation of the ‘Azimuth Looks number’ depending on the ‘number of range looks’ I specify ?

I’m assuming the answer is very simple, it’s even probably in the basics of multilooking but as I said I’m new in SAR techniques (my main work fields are the thermal infrared data from geostationnary satellites).

Thanks in advance :slight_smile:

Yannick Guéhenneux

I had a look at the Multilook operator. The default for both parameters is 1.
So as long as you don’t specify any other value, 1 is used.
Maybe @Luis can tell you more about the usage of the Multilook operator.